Partial Mutual Information Analysis of Financial Networks

The econophysics approach to socio-economic systems is based on the assumption of their complexity. Such assumption inevitably leads to another assumption, namely that underlying interconnections within socio-economic systems, particularly financial markets, are nonlinear, which is shown to be true even in mainstream economic literature. Thus it is surprising to see that network analysis of financial markets is based on linear correlation and its derivatives. An analysis based on partial correlation is of particular interest as it leads to the vicinity of causality detection in time series analysis. In this paper we generalise the planar maximally filtered graphs and partial correlation planar graphs to incorporate nonlinearity using partial mutual information